MEGA-Bench Task - Red Teaming Captcha: leaderboard

The "Red Teaming Captcha" subtask of MEGA-Bench Task. A single task from MEGA-Bench, a large-scale multimodal evaluation suite that scores vision-language models across 500+ real-world tasks spanning perception, reasoning, OCR, charts, documents, and multi-image understanding.

Metric: Task Score (%). Source: huggingface.co. 44 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.0 Flash (Preview)13.2#370
2Qwen 2 VL 2B12.1#1220
3InternVL2.5-2B11.1#1249
4GPT-4o10#343
5Gemma 3 27B (IT)10#523
6Gemma 3 12B (IT)10#664
7Gemma 3 4B (IT)10#992
8GPT-4o Mini10#599
9Claude 3.5 Sonnet (20241022)10#291
10InternVL3-8B10#620
11Gemini 1.5 Pro (002)10#349
12Gemini 1.5 Flash (002)10#583
13Claude 3.5 Sonnet (20240620)10#299
14InternVL3-38B10#405
15InternVL3-78B10#352

Interactive version: theaggregate.ai/benchmark?slug=mega-bench-task-red-teaming-captcha · How It Works · Data refreshed daily, snapshot 2026-10-20.